| blob | 7799cec1dddbdd24121885360f77c88c8b97a746 |
1 Git v1.8.3 Release Notes
2 ========================
4 Backward compatibility notes (for Git 2.0)
5 ------------------------------------------
7 When "git push [$there]" does not say what to push, we have used the
8 traditional "matching" semantics so far (all your branches were sent
9 to the remote as long as there already are branches of the same name
10 over there). In Git 2.0, the default will change to the "simple"
11 semantics that pushes the current branch to the branch with the same
12 name, only when the current branch is set to integrate with that
13 remote branch. There is a user preference configuration variable
14 "push.default" to change this. If you are an old-timer who is used
15 to the "matching" semantics, you can set it to "matching" to keep the
16 traditional behaviour. If you want to live in the future early,
17 you can set it to "simple" today without waiting for Git 2.0.
19 When "git add -u" and "git add -A", that does not specify what paths
20 to add on the command line is run from inside a subdirectory, these
21 commands will operate on the entire tree in Git 2.0 for consistency
22 with "git commit -a" and other commands. Because there will be no
23 mechanism to make "git add -u" behave as if "git add -u .", it is
24 important for those who are used to "git add -u" (without pathspec)
25 updating the index only for paths in the current subdirectory to start
26 training their fingers to explicitly say "git add -u ." when they mean
27 it before Git 2.0 comes. A warning is issued when these commands are
28 run without a pathspec and when you have local changes outside the
29 current directory, because the behaviour in Git 2.0 will be different
30 from today's version in such a situation.
33 Updates since v1.8.2
34 --------------------
36 UI, Workflows & Features
38 * "git branch --vv" learned to paint the name of the branch it
39 integrates with in a different color (color.branch.upstream,
40 which defaults to blue).
42 * "git count-objects" learned "--human-readable" aka "-H" option to
43 show various large numbers in Ki/Mi/GiB scaled as necessary.
45 * "git cherry-pick $blob" and "git cherry-pick $tree" are nonsense,
46 and a more readable error message e.g. "can't cherry-pick a tree"
47 is given (we used to say "expected exactly one commit").
49 * The "--annotate" option to "git send-email" can be turned on (or
50 off) by default with sendemail.annotate configuration variable (you
51 can use --no-annotate from the command line to override it).
53 * The "--cover-letter" option to "git format-patch" can be turned on
54 (or off) by default with format.coverLetter configuration
55 variable. By setting it to 'auto', you can turn it on only for a
56 series with two or more patches.
58 * The bash completion support (in contrib/) learned that cherry-pick
59 takes a few more options than it already knew about.
61 * "git help" learned "-g" option to show the list of guides just like
62 list of commands are given with "-a".
64 * A triangular "pull from one place, push to another place" workflow
65 is supported better by new remote.pushdefault (overrides the
66 "origin" thing) and branch.*.pushremote (overrides the
67 branch.*.remote) configuration variables.
69 * "git status" learned to report that you are in the middle of a
70 revert session, just like it does for a cherry-pick and a bisect
71 session.
73 * The handling by "git branch --set-upstream-to" against various forms
74 of erroneous inputs was suboptimal and has been improved.
76 * When the interactive access to git-shell is not enabled, it issues
77 a message meant to help the system administrator to enable it.
78 An explicit way to help the end users who connect to the service by
79 issuing custom messages to refuse such an access has been added.
81 * In addition to the case where the user edits the log message with
82 the "e)dit" option of "am -i", replace the "Applying: this patch"
83 message with the final log message contents after applymsg hook
84 munges it.
86 * "git status" suggests users to look into using --untracked=no option
87 when it takes too long.
89 * "git status" shows a bit more information to "git status" during a
90 rebase/bisect session.
92 * "git fetch" learned to fetch a commit at the tip of an unadvertised
93 ref by specifying a raw object name from the command line when the
94 server side supports this feature.
96 * Output from "git log --graph" works better with submodule log
97 output now.
99 * "git count-objects -v" learned to report leftover temporary
100 packfiles and other garbage in the object store.
102 * A new read-only credential helper (in contrib/) to interact with
103 the .netrc/.authinfo files has been added.
105 * "git send-email" can be used with the credential helper system.
107 * There was no Porcelain way to say "I no longer am interested in
108 this submodule", once you express your interest in a submodule with
109 "submodule init". "submodule deinit" is the way to do so.
111 * "git pull --rebase" learned to pass "-v/-q" options to underlying
112 "git rebase".
114 * The new "--follow-tags" option tells "git push" to push relevant
115 annotated tags when pushing branches out.
117 * "git merge" and "git pull" can optionally be told to inspect and
118 reject when merging a commit that does not carry a trusted GPG
119 signature.
121 * "git mergetool" now feeds files to the "p4merge" backend in the
122 order that matches the p4 convention, where "theirs" is usually
123 shown on the left side, which is the opposite from other backend
124 expects.
126 * "show/log" now honors gpg.program configuration just like other
127 parts of the code that use GnuPG.
129 * "git log" that shows the difference between the parent and the
130 child has been optimized somewhat.
132 * "git difftool" allows the user to write into the temporary files
133 being shown; if the user makes changes to the working tree at the
134 same time, one of the changes has to be lost in such a case, but it
135 tells the user what happened and refrains from overwriting the copy
136 in the working tree.
138 * There was no good way to ask "I have a random string that came from
139 outside world. I want to turn it into a 40-hex object name while
140 making sure such an object exists". A new peeling suffix ^{object}
141 can be used for that purpose, together with "rev-parse --verify".
144 Performance, Internal Implementation, etc.
146 * Updates for building under msvc.
148 * The stack footprint of some codepaths that access an object from a
149 pack has been shrunk.
151 * The logic to coalesce the same lines removed from the parents in
152 the output from "diff -c/--cc" has been updated, but with an O(n^2)
153 complexity, so this might turn out to be undesirable.
155 * The code to enforce permission bits on files in $GIT_DIR/ for
156 shared repositories have been simplified.
158 * A few codepaths knew how much data they need to put in the
159 hashtables they use upfront, but still started from a small table
160 repeatedly growing and rehashing.
162 * The API to walk reflog entries from the latest to older, which was
163 necessary for operations such as "git checkout -", was cumbersome
164 to use correctly and also inefficient.
166 * Codepaths that inspect log-message-to-be and decide when to add a
167 new Signed-off-by line in various commands have been consolidated.
169 * The pkt-line API, implementation and its callers have been cleaned
170 up to make them more robust.
172 * Cygwin port has a faster-but-lying lstat(2) emulation whose
173 incorrectness does not matter in practice except for a few
174 codepaths, and setting permission bits to directories is a codepath
175 that needs to use a more correct one.
177 * "git checkout" had repeated pathspec matches on the same paths,
178 which have been consolidated. Also a bug in "git checkout dir/"
179 that is started from an unmerged index has been fixed.
181 * A few bugfixes to "git rerere" working on corner case merge
182 conflicts have been applied.
185 Also contains various documentation updates and code clean-ups.
188 Fixes since v1.8.2
189 ------------------
191 Unless otherwise noted, all the fixes since v1.8.2 in the maintenance
192 track are contained in this release (see release notes to them for
193 details).
195 * When "upload-pack" fails while generating a pack in response to
196 "git fetch" (or "git clone"), the receiving side mistakenly said
197 there was a programming error to trigger the die handler
198 recursively.
199 (merge 1ece66b jk/a-thread-only-dies-once later to maint).
201 * "rev-list --stdin" and friends kept bogus pointers into input
202 buffer around as human readble object names. This was not a huge
203 problem but was exposed by a new change that uses these names in
204 error output.
205 (merge 70d26c6 tr/copy-revisions-from-stdin later to maint).
207 * Smart-capable HTTP servers were not restricted via the
208 GIT_NAMESPACE mechanism when talking with commit-walker clients,
209 like they do when talking with smart HTTP clients.
210 (merge 6130f86 jk/http-dumb-namespaces later to maint).
212 * "git merge-tree" did not omit a merge result that is identical to
213 "our" side in certain cases.
214 (merge aacecc3 jk/merge-tree-added-identically later to maint).
216 * Perl scripts like "git-svn" closed (not redirecting to /dev/null)
217 the standard error stream, which is not a very smart thing to do.
218 Later open may return file descriptor #2 for unrelated purpose, and
219 error reporting code may write into them.
220 (merge a749c0b tr/perl-keep-stderr-open later to maint).
222 * "git show-branch" was not prepared to show a very long run of
223 ancestor operators e.g. foobar^2~2^2^2^2...^2~4 correctly.
224 (merge aaa07e3 jk/show-branch-strbuf later to maint).
226 * "git diff --diff-algorithm algo" is also understood as "git diff
227 --diff-algorithm=algo".
228 (merge 0895c6d jk/diff-algo-finishing-touches later to maint).
230 * The new core.commentchar configuration was not applied to a few
231 places.
232 (merge 89c3bbd rt/commentchar-fmt-merge-msg later to maint).
234 * "git bundle" did not like a bundle created using a commit without
235 any message as its one of the prerequistes.
236 (merge 5446e33 lf/bundle-with-tip-wo-message later to maint).
238 * "git log -S/-G" started paying attention to textconv filter, but
239 there was no way to disable this. Make it honor --no-textconv
240 option.
241 (merge 61690bf sr/log-SG-no-textconv later to maint).
243 * When used with "-d temporary-directory" option, "git filter-branch"
244 failed to come back to the original working tree to perform the
245 final clean-up procedure.
246 (merge 9727601 jk/filter-branch-come-back-to-original later to maint).
248 * "git merge $(git rev-parse v1.8.2)" behaved quite differently from
249 "git merge v1.8.2", as if v1.8.2 were written as v1.8.2^0 and did
250 not pay much attention to the annotated tag payload. Make the code
251 notice the type of the tag object, in addition to the dwim_ref()
252 based classification the current code uses (i.e. the name appears
253 in refs/tags/) to decide when to special case merging of tags.
254 (merge a38d3d7 jc/merge-tag-object later to maint).
256 * Fix 1.8.1.x regression that stopped matching "dir" (without
257 trailing slash) to a directory "dir".
258 (merge efa5f82 jc/directory-attrs-regression-fix later to maint-1.8.1).
260 * "git apply --whitespace=fix" was not prepared to see a line getting
261 longer after fixing whitespaces (e.g. tab-in-indent aka Python).
262 (merge 329b26e jc/apply-ws-fix-tab-in-indent later to maint-1.8.1).
264 * The prompt string generator (in contrib/completion/) did not notice
265 when we are in a middle of a "git revert" session.
266 (merge 3ee4452 rr/prompt-revert-head later to maint).
268 * "submodule summary --summary-limit" option did not support
269 "--option=value" form.
270 (merge 862ae6c rs/submodule-summary-limit later to maint).
272 * "index-pack --fix-thin" used an uninitialized value to compute
273 delta depths of objects it appends to the resulting pack.
274 (merge 57165db jk/index-pack-correct-depth-fix later to maint).
276 * "index-pack --verify-stat" used a few counters outside protection
277 of mutex, possibly showing incorrect numbers.
278 (merge 8f82aad nd/index-pack-threaded-fixes later to maint).
280 * The code to keep track of what directory names are known to Git on
281 platforms with case insensitive filesystems can get confused upon a
282 hash collision between these pathnames and looped forever.
284 * Annotated tags outside refs/tags/ hierarchy were not advertised
285 correctly to the ls-remote and fetch with recent version of Git.
287 * Recent optimization broke shallow clones.
288 (merge f59de5d jk/peel-ref later to maint).
290 * "git cmd -- ':(top'" was not diagnosed as an invalid syntax, and
291 instead the parser kept reading beyond the end of the string.
293 * "git tag -f <tag>" always said "Updated tag '<tag>'" even when
294 creating a new tag (i.e. not overwriting nor updating).
296 * "git p4" did not behave well when the path to the root of the P4
297 client was not its real path.
298 (merge bbd8486 pw/p4-symlinked-root later to maint).
300 * "git archive" reports a failure when asked to create an archive out
301 of an empty tree. It would be more intuitive to give an empty
302 archive back in such a case.
304 * When "format-patch" quoted a non-ascii strings on the header files,
305 it incorrectly applied rfc2047 and chopped a single character in
306 the middle of it.
308 * An aliased command spawned from a bare repository that does not say
309 it is bare with "core.bare = yes" is treated as non-bare by mistake.
311 * In "git reflog expire", REACHABLE bit was not cleared from the
312 correct objects.
314 * The logic used by "git diff -M --stat" to shorten the names of
315 files before and after a rename did not work correctly when the
316 common prefix and suffix between the two filenames overlapped.
318 * The "--match=<pattern>" option of "git describe", when used with
319 "--all" to allow refs that are not annotated tags to be used as a
320 base of description, did not restrict the output from the command
321 to those that match the given pattern.
323 * Clarify in the documentation "what" gets pushed to "where" when the
324 command line to "git push" does not say these explicitly.
326 * The "--color=<when>" argument to the commands in the diff family
327 was described poorly.
329 * The arguments given to pre-rebase hook were not documented.
331 * The v4 index format was not documented.
333 * The "--match=<pattern>" argument "git describe" takes uses glob
334 pattern but it wasn't obvious from the documentation.
336 * Some sources failed to compile on systems that lack NI_MAXHOST in
337 their system header (e.g. z/OS).
339 * Add an example use of "--env-filter" in "filter-branch"
340 documentation.
342 * "git bundle verify" did not say "records a complete history" for a
343 bundle that does not have any prerequisites.
345 * In the v1.8.0 era, we changed symbols that do not have to be global
346 to file scope static, but a few functions in graph.c were used by
347 CGit from sideways bypassing the entry points of the API the
348 in-tree users use.
350 * "git update-index -h" did not do the usual "-h(elp)" thing.
352 * "git index-pack" had a buffer-overflow while preparing an
353 informational message when the translated version of it was too
354 long.
356 * 'git commit -m "$msg"' used to add an extra newline even when
357 $msg already ended with one.
359 * The SSL peer verification done by "git imap-send" did not ask for
360 Server Name Indication (RFC 4366), failing to connect SSL/TLS
361 sites that serve multiple hostnames on a single IP.
363 * perl/Git.pm::cat_blob slurped everything in core only to write it
364 out to a file descriptor, which was not a very smart thing to do.
366 * "git branch" did not bother to check nonsense command line
367 parameters and issue errors in many cases.
369 * Verification of signed tags were not done correctly when not in C
370 or en/US locale.
372 * Some platforms and users spell UTF-8 differently; retry with the
373 most official "UTF-8" when the system does not understand the
374 user-supplied encoding name that are the common alternative
375 spellings of UTF-8.
377 * When export-subst is used, "zip" output recorded incorrect
378 size of the file.
380 * "git am $maildir/" applied messages in an unexpected order; sort
381 filenames read from the maildir/ in a way that is more likely to
382 sort messages in the order the writing MUA meant to, by sorting
383 numeric segment in numeric order and non-numeric segment in
384 alphabetical order.
386 * "git submodule update", when recursed into sub-submodules, did not
387 accumulate the prefix paths.